Lecture 3 4. prnet

Chandra Prakash

LPU

Origins of Ad Hoc: Packet Radio

Networks

1

Objectives of the Chapter

Introduction PRNETs

Architecture of PRNETs

Components of Packet Radios

Routing in PRNETs

Route Calculation

Pacing Techniques

Media Access in PRNETs

Flow Acknowledgments in PRNETs

Technical Challenges

2

Ad hoc networks Temporary network composed of mobile nodes without preexisting

communication infrastructure, such as Access Point (AP) and Base Station (BS).

Each node plays the role of router for multi-hop routing.

Self-organizing network without infrastructure networks

Started from DARPA PRNet in 1970

Cooperative nodes (wireless)

Each node decode-and-forward packets for other nodes

Multi-hop packet forwarding through wireless links

Proactive/reactive/hybrid routing protocols

Most works based on CSMA/CA to solve the interference problem

IEEE 802.11 MAC

3 Chandra Prakash, LPU

History

Principle behind ad hc networking ::: Multi hop relaying

(In 522-486 B.C) King of Persia –use line of shouting

men positioned on tall structures or heights.

Video

4

5

Origin of Ad-Hoc Networks

Generic View of mobile ad hoc networks

First generation

They were used for different military scenarios.

Packet radio networks was the first ad-hoc network system

Second generation from 1980s’ to the mid 1990’s

Main aim were the same as for the first generation ad-hoc networks

system i.e.Aiding combat/Battle operations.

Second generation developments focused on the further advancement of

the previously build ad-hoc network structure.

Some important developments: Global mobile information

Systems, Near term Digital Radio (NTDR)

Third generation ad-hoc network systems

known as commercial ad-hoc network systems.

developments , Bluetooth – ad-hoc sensor networks etc6

First generation ad-hoc network

systems

The merits of having an infrastructure-less network were discovered in

the 1970s.

At that time, computers were bulky and so were radio transceivers.

Defence Advanced Research Projects Agency (DARPA) had a

project known as packet radio, where several wireless terminals could

communicate with one another on a battlefield.

DARPA initiated research on the feasibility of using packet-switched

radio communications to provide reliable computer communications.

Came up with packet radio network 1973-1987

Packet radio was a technology that extended the concept of packet

switching (which evolved from point-to-point communication

networks) to the domain of broadcast radio networks.

7

Cont… During the 1970s, the ALOHA project at the University of Hawaii

demonstrated the feasibility of using the broadcasting property of radios to send/receive data packets in a single radio hop system.

ALOHA net utilized single-hop wireless packet switching and a multiple access solution for sharing a single channel.

The ALOHA project later led to the development of a multi-hop multiple-access packet radio network (PRNET).

Initial attempt had a centralized control.

Unlike ALOHA, PRNET permits multi-hop communications over a wide geographical area.

The DARPA PRNET has evolved through the years (1973-1987) to be a robust, reliable, operational experimental network.

The DARPA PRNET projects includes network devices, routing protocols and protocols for automatic distributed network management.

8

Features of PRNET One of the most attractive features of PRNET is rapid

deployment.

Once installed, the system is self-initializing and self-organizing.

Network nodes should be able to discover radio connectivity

among neighboring nodes and organize routing strategies based

on this connectivity.

PRNETs are expected to require no system administration and

can be left unattended.

9

First generation ad-hoc network

systems

Packet radio network components

Firmware

Firmware can be loaded into a Packet Radio(PR) either locally (via serial

interface) or from the PRNET.

The firmware in each PR gathers information about bidirectional link quality,

nodal capacity and route characteristics and provides this knowledge to

debugging and monitoring

Communication

Use radio frequency technology to transmit and receive data

The implemented packet radios support omni-directional, spread spectrum,

half-duplex transmission and reception at 400kbit/s and 100kbit/s rates

They implement the physical, data link and network layer (OSI model).

10

Second generations ad-hoc network

systems

Started in 1980’s -1993

Main Aim

Providing packet switched networking to the mobile battlefield elements in

infrastructure- less environments.

Beneficial in improving

Radios performance by making them smaller, cheaper and power-thrifty.

Scalability of algorithms

Resilience to the electronic attacks.

Survivable radio network (SURAN)– aimed at providing adhoc n/w

with small, low cost, low power, devices with efficient protocols.

Internet Engineering Task Force (IETF) – introduce MANET term

-Work to standardized the protocol and functional specifications of ad hoc network. 11

Second generations ad-hoc network

systems

GloMo (Global Mobile Information Systems) project

Aim: to make the mobile environment user friendly connectivity and access

to services for wireless mobile users.

Includes self organizing/self healing networks; both flat and hierarchical

multihop routing algorithms

12

Second generations ad-hoc network

systems

NearTerm Digital Radio Systems

The NTDR system is an experimental, mobile packet data radio

network.

The NTDR provides a self-organizing, self-healing, network

capability. Radio network management is provided by a Network

ManagementTerminal.

The primary purpose of the NTDR is to provide data transport for

the Army Battle Command System automated systems to units at

brigade and below

Lessons learned from this experimental fielding provide a portion of

the technical baseline for radios being designed for future fielding13

Third generation mobile ad hoc

network system

1990’s- onwards

known as commercial ad-hoc network systems.

Invention of notebook computers and viable communication devices

based on radio waves concept of commercial ad-hoc networks has

arrived.

Idea of collection of mobile nodes were purposed in research

conferences

two main and important applications of mobile ad-hoc networks

Bluetooth – proposed by Ericsson in 1994

Ad-hoc sensors

14

Network Architecture of PRNETs

15

PRNET consists of several

Mobile radio repeaters:

The role of a repeater is to relay packets from one repeater to another, until the packets eventually reach the destination host.

Wireless terminals

Dedicated Mobile stations :

The mobile station is present to derive routes from one host to another.

has knowledge of the overall network connectivity, that is, the network topology.

As network conditions change (terminal movement, repeater failures or recovery, changes in hop reliability, and network congestion state), routes are dynamically reassigned by the station to satisfy minimum delay criteria.

Hosts and terminals attached to the PRNET are unaware of the station's assignment and reassignment of communication routes.

Network Architecture of PRNETs

16

Mobile radio repeaters:

Wireless terminalsMobile stations

Components of Packet Radios

User computer -> mobile device/terminal

User computer is interfaced to a radio via the terminal-network controller (TNC).

Radio and TNC logic ->packet radio

17

Components of Packet Radios

18

The packet radio, therefore, implements functions related to

protocol layers 1, 2, and 3.

It is an intermediate system (IS) in the ISO context.

A packet radio network (PRN) is a collection of packet radios,

with some packet radios connected to user devices while

others are not.

Routing in PRNETs

1. Point-to-Point Routing

PRNETs support point-to-point communications through point-to-point routing.

A packet originating at one part of the network moves through a series of one or

more repeaters until it eventually reaches the final destination.

Point-to-point route is an ordered set of repeater addresses

Specific routes are derived prior to data transmission

determined by the mobile station:

Mobile station is the only element in the network that has knowledge of

the overall network connectivity (the network topology).

Mobile station computes the best point-to-point route and distributes this information

to all repeaters in the route or directly to the source packet radio.

This scheme was found to be suitable for slow moving user terminals.

19

Routing in PRNETs

2. Broadcast Routing

Radio technology provides very good broadcasting properties.

Broadcasting information to all radios in a network is equivalent to

flooding.

To ensure that each packet radio only forwards a packet once, each

repeater has to maintain a list of packet identifiers for previously

broadcast packets that it recently had received and forwarded.

In broadcast routing, a packet radiates away from the source packet

radio in a wave-like fashion, i.e the packet ripples away from the

source.

20

Routing in PRNETs

2. Broadcast Routing (cont…)

Broadcasting is very robust

All other nodes in the network must participate in the transmission and

reception of packets even that is not intended for them.

When broadcast routing is used for point-to-point communication, the

destination host address is included in each data packet.

Routing decisions are not centralized:

No specific routes are derived prior to data transmission

Packets will eventually reach the destination host if the network is not

partitioned. For fast moving user terminals, broadcast routing was

found to be useful as it avoids the need to process rapidly changing

routes.21

Packet Forwarding1. Connectionless approach :

Requires some background operation to maintain up-to-date network

topology and link information in each node.

As network topology changes, the background routing traffic can be

substantial.

Associated with broadcast routing, where each packet carries sufficient

routing information for it to arrive at the destination.

2. Connection-oriented packet : An Explicit route establishment phase is required before data traffic can

be transported.

Associated with point-to-point routing, where each node in a route

has a lookup table for forwarding incoming packets to the respective

outgoing links.

If a topology changes, a route re-establishment phase is needed.22

Impact of Mobility In a PRNET, all elements of the network can be mobile.

Assumption : User terminals normally move slowly enough such that the assigned

point-to-point routes are valid for at least a few seconds before another route must

be chosen.

If User rate of mobility increased point-to-point routing may not

be practical . Why ???

Because most of the time will be spent in computing alternate point-to-point

routes instead of forwarding the packets to their intended destinations.

Solution ???

Broadcast routing is less affected by user mobility

the packets do not follow a specific point-to-point route.

Every node is supposed to relay the packets to the destination host.

No need to cope with rapidly changing routes in broadcast routing under

conditions of rapid host mobility.

broadcasting is power inefficient.23

Route CalculationEach packet radio gathers and maintains information about current

network topology so that it can make independent decisions about

how to route packets toward their destinations.

Each node maintains the following tables:

1. Neighbor table

2. Tier table

3. Device table

24

Neighbor Table

Broadcast a Packet Radio Organization Packet (PROP)

every 7.5 seconds

Announcing its existence and information about the

network topology from its own perspective at time.

Neighbors that hear a PROP make entry in their neighbor tables

When nodes hears a PROP, it updates its neighbor table

Transmitted data packets also used to build neighbor table

Tracks bidirectional quality of links with neighbors

(retransmission counts)

25

Neighbor Table

Neighboring PR Link Quality

Node 1 3/9

Node 5 4/5

Node 7 6/9

Node 9 5/8

26

Link Quality =

number of packets correctly received from the transmitting packet radio

during a PROP

number of packets that the transmitting packet radio actually transmitted at

that same interval

Tier Table Routing in PRNETs relies on each packet radio maintaining

adequate knowledge of the best packet radio to forward

packets to for every prospective destination.

The tier information ripples outward from each packet radio at an

average rate of 3.75 seconds per hop and eventually reaches all

packet radios.

Every packet radio knows its distance in tiers (or radio hops) from

itself to every prospective destination and the next-hop packet

radio.

Similar to the early ARPANET routing algorithm which is based

on the classical Bellman-Ford routing.

27

Tier Table

Every packet radio knows the “best” next node on the route from it to a given

destination node

Tier 1 = 1 hop neighbors

These neighbors send out their PROPs indicating that they are one hop from

the originator

At next step, receivers of these PROPs know that they are 2 hops away from

the originator

Process continues until every radio knows its distance in tiers from every

other radio

“Best”: shortest route with “good” connectivity on each hope

To change table, must discover a new node with better link quality and

lower tier number than currently recorded next node

Also disseminate information about bad links in PROP messages

28

Tier Table

Destination PR Next-Hop PR Tier Count

Node 1 Node 7 2

Node 4 Node 4 0

Node 5 Node 8 1

Node 6 Node 3 1

29

Device Table

30

Each mobile device/terminal periodically sends a

control packet across the wired interface to its

attached packet radio.

Packet radio keeps track of affiliated devices and

propagates this mapping information via a PROP to

other packet radios in the network at an average rate

of 3.75 seconds per hop.

when a packet radio receives a packet addressed to a

specific mobile device, the device currently attached

to the packet radio is known and the appropriate next

hop-packet radio is chosen to forward the packet.

Device Table

Logical addressing: maps device to a packet radio

Information about the radio’s attached device is included in PROP messages

This allows new radios to be attached to devices and vice versa

Such correspondences are maintained in the device table at each packet radio

P

NML

Q

1 2

Device

PR Node

Device

31

Principles of Forwarding ProtocolsForwarding is accomplished via information read from the device and tier tables and from

the packet headers.

Packet Headers:

End-to-end header

The end-to-end header (ETE) is created by the source mobile

device/terminal, not the packet radio.

<Src Device ID, Dest Device ID, Type of Service Flag>

Source device ID/address: used to update the packet radio's device-to-

packet radio mapping information

Destination device ID/address: used in packet forwarding.

ToS: indicate low latency/low reliability, e.g., speech

The ETE header remains intact as the packet transits toward the destination

device.32

Principles of Forwarding Protocols

Routing header

In contrast to the ETE header, the routing header is created by the

source packet radio.

The routing header encapsulates the ETE header, since it is the routing

header that the packet radio will use to forward the packets

<Src PR ID, Seq No, Speech ToS flag, Prev PR ID (for acks), Prev PR

transmission count, Transmitting PR ID, Transmitting PR transmit count (for

pacing), Next PR ID, Lateral alternative routing flag, Alternative routing

request flag, Tier, Dest PR ID>

33

Routing Header

34

Forwarding Protocol

Device 1 --> Device 2 via PRs L, M, N

Device 1 --> PR L

Device sends packet PR L via its wired connection;

Prepare packet to forward on to PR N via PR M:

Dest PR ID =N

Prev PR ID =null

Trans PR ID = L

Next PR ID =M (known from tier table)

Tier =2 (from tier table)

P

NML

Q

1 2

Device

PR Node

Device

35

Forwarding Protocol

PR L --> PR M

PR M receives packet over the air

Next PR ID = M, this PR should process the packet

Prepare to forward packet on to PR N:

Prev PR ID = L

Transmitting PR ID = M

Next PR ID = N (known from tier table)

Tier = 1 (from tier table)

Transmit packet to PR N … and any other PR within range, including L! This is an

example of the passive acknowledgement.

P

NML

Q

1 2

Device

PR Node

Device

36

Forwarding Protocol

PR M --> PR N

N receives packet, determines it should process it based on Next PR ID

Determines that packet should be delivered to the attached Device 2 (from ETE header

and device table)

Wire-line transmits the packet to Device 2

Sets in header, for the ack message:

Prev PR ID = M

Trans PR ID = N

Next PR ID = null

Tier = null

Ack message is sent, consisting only of header

Note that end PR can’t use passive acknowledgement, so is forced to transmit ack

message to PR M

P

NML

Q

1 2

Device

PR Node

Device

37

Forwarding Protocol Criteria for recognizing an Ack

Source PR ID and Seq No match the original packet

AND must have arrived from further downstream:

Transmitting PR ID in ack packet is same as next PR ID in original

packet

Previous PR ID is same as receiving PR’s ID--the forward packet

came from this packet radio

Ack packet contains a smaller tier number, indicating it got closer

to the destination PR

38

Forwarding Protocol Retransmissions

If a packet is forwarded, and no ack is received, the packet will be

retransmitted after a time out

Will do this six times before giving up

Interval between retransmission based on pacing protocol, and

grows with each successive unsuccessful retransmission

At some point, sending PR assumes that it can no longer reach the

next radio on the hop and sets its connectivity to that radio to 0

39

Transmission Protocols

Pacing protocol

Provide Flow and congestion control mechanisms

Transmission parameters are chosen based on measure link

quality and the type of service desired by the user.

Also promotes fair use of the radio spectrum

The time at which a packet is selected for transmission is

determined by a three-component packing protocol.

Single Threading

Forwarding Delay Measurement

Measurement of Retransmissions

40

Transmission Protocols Single Threading

Last packet sent to PR must be ack’d or discarded before next packet is sent

to the same PR

Passive acks imply that next hop PR now ready to accept a new packet

Deflects congestion bottleneck away from source PR

Forwarding Delay

packet radio records the time at which its transmission completes and when it

receives the acknowledgment from the next packet radio. This difference is

known as forwarding delay

Affects the setting of retransmission intervals

Includes processing, queuing, carrier sense/random access, transmission

delay from neighboring PR

41

42

Media Access in PRNET’s

PRNETs employ the Carrier Sense Multiple Access (CSMA)

protocol to coordinate communications among mobile hosts.

CSMA prevents a packet radio from transmitting at the same

time when a neighboring packet radio is using the medium.

A packet radio is aware if a neighbor is transmitting by

reading its hardware indication bit-synchronization-in-the-

lock.

Basically, this bit, when set, implies that the channel is busy

and a carrier is being sensed.

Whenever a carrier is being sensed, a packet radio will

refrain from transmitting.

43

CSMA Carrier Sense Multiple Access (CSMA) is a probabilistic Media Access

Control (MAC) protocol in which a node verifies the absence of

other traffic before transmitting on a shared transmission medium, such as an electrical

bus, or a band of the electromagnetic spectrum.

"Carrier Sense" describes the fact that a transmitter uses feedback from a receiver that

detects a carrier wave before trying to send. That is, it tries to detect the presence of an

encoded signal from another station before attempting to transmit. If a carrier is

sensed, the station waits for the transmission in progress to finish before initiating its

own transmission. In other words, CSMA is based on the principle "sense

before transmit" or "listen before talk".

"Multiple Access" describes the fact that multiple stations send and receive on the

medium. Transmissions by one node are generally received by all other stations using

the medium.

44

Media Access in PRNET’s

While carrier sensing reduces the probability of channel

contention, it cannot eliminate hidden terminal and

exposed nodes problems.

The former is a result of a node lying within the radio range of

the receiver, but not another transmitter that is two hops away.

The latter results in neighboring nodes of a transmitter being

blocked from transmission.

45

Flow Acknowledgments in PRNETs

Packets are forwarded via a single communication route

through a PRNET.

Each packet radio must

examines the information contained in the packet headers and

in its own device and tier tables.

decide if it should be the one to transmit the packet, if it should

update the routing header before transmitting, and if it should

update its own tables.

46

47

Other packet radios within the radio range will also receive the

transmitted packet.

If these neigbors are not part of the route, they will discard the

overheard packets.

The downstream node that receives the packet will process the

packet and proceed with issuing a passive acknowledgment.

The single transmission, not only forwards the packet on to the

next packet radio but also acknowledges the previous packet radio

that the packet was successfully received and is being forwarded.

This principle of passive acknowledgment will proceed

until the packet reaches the destination node. reception of the

packet.

48

Since the destination node does not have a downstream

node and it is the terminating point, an active

acknowledgment is sent by the destination node to its

upstream node to confirm successful

49

Technical Challenges

PRNETs are different from wired networks in many aspects. They have an infrastructure-less backbone and network nodes that act as

routers or packet switches to forward packets from one node to another.

Routers are connected without wires and routers themselves can be mobile. The introduction of wireless connectivity and the presence of mobility

result in great technical challenges in the field of computer communications.

PRNET is the network that attempts to merge computer communications with telecommunications. It allows networks to be formed and de-formed on-the-fly, through a set

of innovative and adaptive communication protocols.

50

Technical challenges for PRNET

The technical challenges for PRNET can be summarized as:

Flow control over a wireless multi-hop communication route

Error control over wireless links

Deriving and maintaining network topology information

Deriving accurate routing information

Mechanisms to handle router mobility

Shared channel access by multiple users

Processing capability of terminals

Size and power requirements

51